The answer is two fold depending on the exact details of the workflow you want to follow in creating the profile.
If the mapping between the source color space and the target color space has been measured and is available, you can use that information to create a new color profile struct which can be passed to "iccwrite". If that is not the case, then Image Processing Toolbox does not offer out of the box functionality for the actual measurement part of the process.
Otherwise, create a new ICC profile by providing all of the necessary values needed for an ICC profile in a "profile struct":
Usually the workflow would involve reading an existing profile for a monitor using "iccread", modifying the tag table, white point values, etc. and then saving the new profile as a ".icm" file using "iccwrite".
Image Processing Toolbox offers functionality to read, write and modify ICC profiles. You can access an example ICC profile using the "iccread" command.
>> inProfile = iccread('monitor.icm')
You can access the the tag tables from the "inProfile" struct itself:
>> inProfile.TagTable
ans =
11×3 cell array
{'desc'} {[ 264]} {[ 97]}
{'rXYZ'} {[ 361]} {[ 20]}
{'gXYZ'} {[ 384]} {[ 20]}
You can modify the tag table by directly change any value within the struct. Eg. - to change the value under row "rXYZ" from 361 to 200, execute:
>> inProfile.TagTable(2, 2) = num2cell(300);
Afterwards, you can write the profile to disk using "iccwrite":
>> outProfile = iccwrite(inProfile, "filename.icm");
This will save the ICC profile to your disk which you can re-use at a later time. The location the profile is saved to is determined by "iccroot".
More information on ICC profile functionality in Image Processing Toolbox can be found here:
Best Answer